Compare

Recraft V3 vs Nemotron-3-120b-A12b

Pricing per million tokens, context window, capabilities — pulled from each provider's public docs. All 2 are available via the same AIgateway OpenAI-compatible endpoint; flip the model string to switch.

Search2/4
Recraft V3
recraft/recraftv3
Nemotron-3-120b-A12b
nvidia/nemotron-3-120b-a12b
Provider
Recraft
Nvidia
Family
Recraft
Modality
image
text
Context window
131,072 tok
Max output
4,096 tok
Released
2026-05-17
2026-02-24
License
Proprietary
Open-weight
Input price
$0.500 /1M
Output price
$1.20 /1M
Per image
$0.040 /img
Tools
yes
Streaming
yes
Vision
JSON mode
yes
Reasoning
yes
Prompt caching
Batch API
Try it
Open in playground →
Open in playground →
Recraft V3
recraft/recraftv3
Full spec →

Recraft V3 is the previous-generation text-to-image model from Recraft, well-suited to design-quality compositions, brand-aware imagery, and accurate text rendering.

Strengths
  • Text-to-image generation
  • Creative control
Use cases
Marketing assetsProduct mockupsConcept art
Nemotron-3-120b-A12b
nvidia/nemotron-3-120b-a12b
Full spec →

NVIDIA Nemotron 3 Super is a hybrid MoE model with leading accuracy for multi-agent applications and specialized agentic AI systems.

Strengths
  • General-purpose chat
  • Long context
  • Tool use
Use cases
ChatbotsContent generationAgentic workflows

Compare with another

GPT-5.5 Pro vs Recraft V3
openai/gpt-5.5-pro · recraft/recraftv3
GPT-5.5 Pro vs Nemotron-3-120b-A12b
openai/gpt-5.5-pro · nvidia/nemotron-3-120b-a12b
GPT-5.5 vs Recraft V3
openai/gpt-5.5 · recraft/recraftv3
GPT-5.5 vs Nemotron-3-120b-A12b
openai/gpt-5.5 · nvidia/nemotron-3-120b-a12b
Claude Opus 4.7 vs Recraft V3
anthropic/claude-opus-4.7 · recraft/recraftv3
Claude Opus 4.7 vs Nemotron-3-120b-A12b
anthropic/claude-opus-4.7 · nvidia/nemotron-3-120b-a12b
Claude Sonnet 4.6 vs Recraft V3
anthropic/claude-sonnet-4.6 · recraft/recraftv3
Claude Sonnet 4.6 vs Nemotron-3-120b-A12b
anthropic/claude-sonnet-4.6 · nvidia/nemotron-3-120b-a12b
Grok 4 vs Recraft V3
xai/grok-4 · recraft/recraftv3
SWITCH BETWEEN THEM

One key, all 2, one line different.

from openai import OpenAI

client = OpenAI(
    base_url="https://api.aigateway.sh/v1",
    api_key="sk-aig-...",
)

# Recraft V3
client.chat.completions.create(
    model="recraft/recraftv3",
    messages=[{"role":"user","content":"hello"}],
)

# Nemotron-3-120b-A12b
client.chat.completions.create(
    model="nvidia/nemotron-3-120b-a12b",
    messages=[{"role":"user","content":"hello"}],
)
Get an AIgateway keyRun an eval on these →